Nanoscale mapping of stacking-dependent work function and local photoresponse in CVD-grown MoS2 bilayers by KPFM

This study utilizes Kelvin Probe Force Microscopy to map the stacking-dependent work function and local photoresponse of CVD-grown MoS2 bilayers, revealing how interlayer coupling, substrate-induced photogating, and surface particulate trapping collectively govern their nanoscale optoelectronic behavior.

The Gist

Imagine you are a tiny architect trying to build a super-efficient solar panel or a lightning-fast computer chip. The materials you are using are called MoS2 (Molybdenum Disulfide). Think of MoS2 not as a solid block, but as a stack of incredibly thin, sticky sheets of paper.

This paper is about how the way you stack these sheets, and the "dirt" left over from the manufacturing process, changes how the material behaves when light hits it.

Here is the story of their discovery, broken down into simple concepts:

1. The Two Ways to Stack the Sheets

Imagine you have two sheets of paper. You can stack them in two main ways:

• The "Perfectly Aligned" Stack (AB-stacking): Like placing one sheet directly on top of another, but slightly shifted so the atoms line up in a specific, tight pattern. This is like a firm handshake between the layers.
• The "Rotated" Stack (AA'-stacking): Like placing one sheet on top of another but rotated slightly, so the atoms don't quite line up perfectly. This is a looser connection.

The researchers found that the "Perfectly Aligned" stack (AB) holds onto its electrons much tighter than the "Rotated" stack. This changes a property called the Work Function.

• The Analogy: Think of the Work Function as the "effort required to pull a magnet off a fridge."
  • In the AB-stack, the magnet is stuck very tight (high effort/work function).
  • In the AA'-stack, it's a bit looser (lower effort/work function).
  • Why? Because the AB-stack layers are holding hands so tightly that they change how easily electricity can flow out of them.

2. The "Dirt" Problem (The NaCl Residue)

To grow these tiny sheets, the scientists used a special recipe involving salt (NaCl). Think of this like using a special glue to stick the sheets to the table.

• The Issue: When the glue dries, it leaves behind tiny, invisible crumbs of salt and other gunk on the surface of the sheets.
• The Effect: These crumbs act like tiny traps. When light hits the material, the electrons get excited and want to run around, but they get caught in these "traps" left by the salt.
• The Map: The scientists used a super-sensitive tool called KPFM (think of it as a microscopic metal detector) to map the surface. They saw "stripes" and "spots" where the electrical voltage was different. These weren't physical bumps you could see with a normal microscope; they were electrical ghosts caused by the salt crumbs.

3. Shining a Light (The Photoresponse)

The big question was: What happens when we shine a light on these sheets?

• The Normal Reaction: Usually, when light hits MoS2, it creates extra electricity (like a solar panel). The material gets "n-doped," which is a fancy way of saying it becomes better at conducting negative charges.
• The Twist: The researchers found that the "dirt" (salt crumbs) changes the story.
  • In some areas, the light makes the material work better because the trapped holes (positive charges) act like a gate, letting more electrons flow.
  • In other areas, the light gets trapped by the dirt, causing weird electrical shifts.
• The Analogy: Imagine a highway (the MoS2 sheet).
  • Without light: Traffic is slow.
  • With light: The traffic lights turn green, and cars zoom.
  • With the "dirt": The salt crumbs are like potholes or construction zones. Sometimes they help traffic flow by clearing a lane (photogating), but sometimes they cause traffic jams (carrier trapping), making the flow uneven.

4. The "Ghost" Triangles

One of the coolest findings was a "ghost" triangle.

• In the middle of some triangular flakes, there was a faint, triangular shape that showed up on the electrical map but not on the physical map.
• The Explanation: It was like a shadow. The salt crumbs were clustered in the center of the triangle. Even though the surface looked flat, the electrical "shadow" of those crumbs stretched out over the whole triangle, changing how the electricity behaved underneath.

5. Why Does This Matter?

If you want to build a reliable device (like a phone sensor or a solar cell), you need the material to behave the same way everywhere.

• The Problem: The salt residue creates a "patchwork quilt" of electrical behavior. Some spots work great, others are sluggish.
• The Solution: This study teaches us that we can't just look at the material; we have to look at how it was made. If we want perfect devices, we need to either clean up the "dirt" better or learn how to design our circuits to work with these variations.

Summary in a Nutshell

The researchers took a microscopic look at MoS2 sheets grown with a salt-assisted method. They discovered that:

1. How you stack the sheets changes how hard it is to pull electricity out of them.
2. Leftover salt crumbs act like electrical traps, creating a messy, uneven landscape of electricity.
3. Shining light interacts with these crumbs in complex ways, sometimes helping and sometimes hindering the flow of electricity.

This is a crucial lesson for engineers: To build the future of electronics, we need to understand not just the material, but the tiny imperfections left behind by the manufacturing process.

Technical Summary

1. Problem Statement

Transition metal dichalcogenides (TMDs), particularly molybdenum disulfide (MoS₂), are promising for optoelectronic applications due to their tunable electronic properties. While the stacking order (e.g., AB-stacking/2H vs. AA'-stacking/3R) is known to influence crystal symmetry and interlayer coupling, a comprehensive understanding of how stacking order affects the local work function and optoelectronic response at the nanoscale remains limited.

Furthermore, the widely used NaCl-assisted Chemical Vapor Deposition (CVD) method for growing large-area TMDs often leaves behind significant surface residues (Na-derived particulates and adsorbates). These residues are known to degrade device performance (e.g., reducing photoluminescence and carrier mobility), yet their specific impact on the local electronic landscape, carrier trapping, and photoresponse of different stacking configurations has not been fully explored.

2. Methodology

The study employed a multi-modal characterization approach to investigate NaCl-assisted CVD-grown MoS₂ bilayers on SiO₂/Si substrates:

• Sample Growth: MoS₂ was grown using a horizontal tube furnace with NaCl as a growth promoter. The process utilized a vapor-liquid-solid (VLS) mechanism, resulting in both monolayer and bilayer domains with distinct stacking orders (AA' and AB).
• Kelvin Probe Force Microscopy (KPFM): This was the primary technique used to map surface potential ($V_{CPD}$) and derive the work function ($\phi$) with nanoscale resolution.
  • Measurements were conducted in dark conditions and under 633 nm laser illumination (5 mW) to probe photo-induced tunability.
  • Tip calibration was performed using Highly Oriented Pyrolytic Graphite (HOPG).
• Complementary Characterization:
  • Raman & Photoluminescence (PL) Spectroscopy: To confirm layer thickness, stacking order, and optical quality.
  • X-ray Photoelectron Spectroscopy (XPS): To identify chemical states and confirm the presence of residual Sodium (Na).
  • Lateral Force Microscopy (LFM) & Force Modulation Microscopy (FMM): To distinguish between topographic/frictional effects and purely electrostatic variations, probing the nanomechanical response.

3. Key Contributions

• Stacking-Dependent Work Function Mapping: The study provides the first direct nanoscale comparison of work function evolution in AA'-stacked (3R) vs. AB-stacked (2H) MoS₂ bilayers.
• Decoupling Residue Effects: It isolates the electronic impact of NaCl-assisted growth residues, demonstrating how surface particulates create localized carrier traps and electrostatic perturbations.
• Photoresponse Mechanism Elucidation: The paper clarifies the competing mechanisms governing the photoresponse in bilayer MoS₂, specifically the interplay between substrate-induced photogating, interlayer coupling, and surface carrier trapping.
• Correlative Nanomechanical/Electronic Analysis: By combining KPFM with LFM and FMM, the authors distinguish between mechanical heterogeneities (caused by particles) and purely electronic phenomena.

4. Key Results

A. Structural and Chemical Characterization

• Stacking Identification: Raman spectroscopy confirmed the bilayer nature of the flakes (indicated by a peak separation of ~23 cm⁻¹ between $E^1_{2g}$ and $A_{1g}$ modes). XPS confirmed the presence of residual Na (Na 1s peak), linking surface particulates to Na-derived species.
• Morphology: Optical and AFM imaging revealed triangular flakes with distinct inner (bilayer) and outer (monolayer) regions. AFM showed apparent height anomalies due to surface particulates.

B. Work Function and Stacking Order

• Layer Dependence: The work function increased with the number of layers in both stacking configurations.
• Stacking Dependence:
  • AB-stacked (2H): Exhibited a significantly larger work function difference between the bottom and top layers (~420 meV). This is attributed to stronger interlayer coupling, which enhances charge redistribution.
  • AA'-stacked (3R): Showed a smaller work function difference (~120–210 meV) due to weaker interlayer hybridization.
• Surface Particulates: Na-derived particulates at the edges of domains created distinct "stripe-like" patterns in surface potential maps, characterized by higher surface potential (lower work function) due to localized carrier trapping and interfacial dipoles.

C. Photoresponse and Photogating

• General n-type Doping: Under 633 nm illumination, most regions showed an increase in surface potential (decrease in work function), indicating n-type doping. This is driven by photogating: photoexcited holes are trapped at the MoS₂/SiO₂ interface (enhanced by Na⁺ ions), acting as a positive gate that increases electron density on the MoS₂ surface.
• Stacking-Specific Response:
  • In AB-stacked bilayers, the work function difference between layers increased under illumination due to enhanced screening and interlayer carrier redistribution.
  • Anomaly in Region R2: One AB-stacked region showed an increase in work function under light, attributed to the trapping of photogenerated electrons in defect states associated with residues.
• Residue Screening: The high-contrast "stripe" patterns caused by residues faded under illumination. This suggests that photogenerated carriers redistribute laterally to screen the electrostatic potential variations induced by the residue network.

D. Nanomechanical Correlation

• LFM/FMM Findings: The "stripe" features observed in KPFM maps were confirmed to be mechanically distinct domains with higher friction and lower stiffness, correlating with the physical presence of surface particulates.
• Purely Electronic Features: The secondary triangular contrast observed in the center of some bilayers (which did not appear in LFM/FMM maps) was identified as purely electrostatic in origin, likely due to the electrostatic response of the underlying layer to localized charge distributions from surface particles.

5. Significance

This study provides critical insights for the design of reliable TMD-based optoelectronic devices:

1. Device Optimization: It highlights that while stacking order (AB vs. AA') dictates intrinsic electronic properties, growth-induced residues are a dominant factor in local electronic heterogeneity.
2. Performance Reliability: The findings explain why NaCl-assisted CVD samples often suffer from performance degradation; surface particulates act as carrier traps and create non-uniform photoresponse landscapes.
3. Methodological Advance: The work demonstrates that KPFM, when combined with mechanical modes (LFM/FMM), is essential for distinguishing between structural defects and intrinsic electronic variations in 2D materials.
4. Future Applications: Understanding the interplay between interlayer coupling and substrate-induced photogating is vital for engineering high-performance photodetectors and transistors using stacked TMDs.